Experimental News Clipping Site

Tag: manipulation

  • Simon Willison’s Weblog: Quoting Bruce Schneier

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Aug/27/bruce-schneier/#atom-everything Source: Simon Willison’s Weblog Title: Quoting Bruce Schneier Feedly Summary: We simply don’t know to defend against these attacks. We have zero agentic AI systems that are secure against these attacks. Any AI that is working in an adversarial environment—and by this I mean that it may encounter untrusted training data or…

  • Slashdot: Google Improves Gemini AI Image Editing With ‘Nano Banana’ Model

    —

    by

    Kurt Seifried
    in

    Source URL: https://slashdot.org/story/25/08/26/215246/google-improves-gemini-ai-image-editing-with-nano-banana-model?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: Google Improves Gemini AI Image Editing With ‘Nano Banana’ Model Feedly Summary: AI Summary and Description: Yes Summary: Google DeepMind has launched the “nano banana” model (Gemini 2.5 Flash Image), which excels in AI image editing by offering improved consistency in edits. This advancement enhances the practical use cases…

  • The Register: One long sentence is all it takes to make LLMs misbehave

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/08/26/breaking_llms_for_fun/ Source: The Register Title: One long sentence is all it takes to make LLMs misbehave Feedly Summary: Chatbots ignore their guardrails when your grammar sucks, researchers find Security researchers from Palo Alto Networks’ Unit 42 have discovered the key to getting large language model (LLM) chatbots to ignore their guardrails, and it’s…

  • The Register: Honey, I shrunk the image and now I’m pwned

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/08/21/google_gemini_image_scaling_attack/ Source: The Register Title: Honey, I shrunk the image and now I’m pwned Feedly Summary: Google’s Gemini-powered tools tripped up by image-scaling prompt injection Security researchers with Trail of Bits have found that Google Gemini CLI and other production AI systems can be deceived by image scaling attacks, a well-known adversarial challenge…

  • Cloud Blog: Intelligent code conversion: Databricks Spark SQL to BigQuery SQL via Gemini

    —

    by

    Kurt Seifried
    in

    Source URL: https://cloud.google.com/blog/products/data-analytics/automate-sql-translation-databricks-to-bigquery-with-gemini/ Source: Cloud Blog Title: Intelligent code conversion: Databricks Spark SQL to BigQuery SQL via Gemini Feedly Summary: As data platforms evolve and businesses diversify their cloud ecosystems, the need to migrate SQL workloads between engines is becoming increasingly common. Recently, I had the opportunity to work on translating a set of Databricks…

  • Unit 42: Logit-Gap Steering: A New Frontier in Understanding and Probing LLM Safety

    —

    by

    Kurt Seifried
    in

    Source URL: https://unit42.paloaltonetworks.com/logit-gap-steering-impact/ Source: Unit 42 Title: Logit-Gap Steering: A New Frontier in Understanding and Probing LLM Safety Feedly Summary: New research from Unit 42 on logit-gap steering reveals how internal alignment measures can be bypassed, making external AI security vital. The post Logit-Gap Steering: A New Frontier in Understanding and Probing LLM Safety appeared…

  • Slashdot: Microsoft Warns Excel’s New AI Function ‘Can Give Incorrect Responses’ in High-Stakes Scenarios

    —

    by

    Kurt Seifried
    in

    Source URL: https://it.slashdot.org/story/25/08/20/128217/microsoft-warns-excels-new-ai-function-can-give-incorrect-responses-in-high-stakes-scenarios?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: Microsoft Warns Excel’s New AI Function ‘Can Give Incorrect Responses’ in High-Stakes Scenarios Feedly Summary: AI Summary and Description: Yes Summary: Microsoft is testing a new AI feature called COPILOT in Excel that leverages OpenAI’s gpt-4.1-mini model for automating spreadsheet tasks through natural language. While it presents innovative capabilities…

  • Cisco Talos Blog: Russian state-sponsored espionage group Static Tundra compromises unpatched end-of-life network devices

    —

    by

    Kurt Seifried
    in

    Source URL: https://blog.talosintelligence.com/static-tundra/ Source: Cisco Talos Blog Title: Russian state-sponsored espionage group Static Tundra compromises unpatched end-of-life network devices Feedly Summary: A Russian state-sponsored group, Static Tundra, is exploiting an old Cisco IOS vulnerability to compromise unpatched network devices worldwide, targeting key sectors for intelligence gathering. AI Summary and Description: Yes Summary: The text provides…

  • Schneier on Security: Subverting AIOps Systems Through Poisoned Input Data

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.schneier.com/blog/archives/2025/08/subverting-aiops-systems-through-poisoned-input-data.html Source: Schneier on Security Title: Subverting AIOps Systems Through Poisoned Input Data Feedly Summary: In this input integrity attack against an AI system, researchers were able to fool AIOps tools: AIOps refers to the use of LLM-based agents to gather and analyze application telemetry, including system logs, performance metrics, traces, and alerts,…

  • Embrace The Red: Amp Code: Invisible Prompt Injection Fixed by Sourcegraph

    —

    by

    Kurt Seifried
    in

    Source URL: https://embracethered.com/blog/posts/2025/amp-code-fixed-invisible-prompt-injection/ Source: Embrace The Red Title: Amp Code: Invisible Prompt Injection Fixed by Sourcegraph Feedly Summary: In this post we will look at Amp, a coding agent from Sourcegraph. The other day we discussed how invisible instructions impact Google Jules. Turns out that many client applications are vulnerable to these kinds of attacks…